mengram-ai 2.5.0

AI memory layer with autonomous agents. Open-source Mem0 alternative with memory agents, webhooks, and cognitive insights.

Mengram — AI Memory Layer with Autonomous Agents

Open-source memory layer for AI apps. Not just storage — Mengram has autonomous agents that clean, analyze, and find hidden patterns in your knowledge.

Website · Dashboard · API Docs · PyPI · npm

---

Why Mengram?

	Mengram	Mem0
Memory Storage	✅	✅
Semantic Search	✅	✅
Knowledge Graph	✅	✅
Autonomous Agents	✅ Curator, Connector, Digest	❌
Team Shared Memory	✅ Invite codes, privacy controls	❌
AI Reflections	✅ Patterns, insights, behavioral analysis	❌
Webhooks	✅	✅
MCP Server	✅ Claude Desktop, Cursor, Windsurf	✅
Python SDK	✅	✅
JavaScript SDK	✅	✅
Async Mode	✅ Job tracking	✅
Memory Categories	✅ user_id, agent_id, run_id, app_id	✅
Self-hostable	✅	✅
Price	Free	$19-249/mo

Quick Start (60 seconds)

1. Get API key

Sign up at mengram.io — free, no credit card.

2. Install

pip install mengram-ai    # Python
npm install mengram-ai    # JavaScript / TypeScript

3. Connect to Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "mengram": {
      "command": "mengram",
      "args": ["server", "--cloud"],
      "env": {
        "MENGRAM_API_KEY": "your-key-here"
      }
    }
  }
}

Done. Claude now has persistent memory.

Python SDK

from mengram.cloud.client import CloudMemory

m = CloudMemory(api_key="mg-...")

# Add memories (async — returns immediately)
result = m.add([
    {"role": "user", "content": "I deployed Mengram on Railway with PostgreSQL 15"},
    {"role": "assistant", "content": "Great, noted the deployment setup."}
], user_id="ali", agent_id="devops-bot", app_id="infra")

# Track background job
job = m.wait_for_job(result["job_id"])

# Semantic search (filter by agent)
results = m.search("deployment setup", user_id="ali", agent_id="devops-bot")

# Run memory agents
m.run_agents(agent="all", auto_fix=True)

# Get AI insights
insights = m.insights()

# Team memory
team = m.create_team("Backend Team")
m.share_memory("Redis", team_id=team["id"])

# Webhooks
m.create_webhook(url="https://hooks.slack.com/...", name="Slack")

# Manage API keys
m.create_key(name="production")
keys = m.list_keys()

JavaScript / TypeScript SDK

const { MengramClient } = require('mengram-ai');

const m = new MengramClient('mg-...');

// Add memories with categories
const { job_id } = await m.add([
  { role: 'user', content: 'I prefer dark mode and use Vim' },
  { role: 'assistant', content: 'Noted your preferences!' }
], { userId: 'ali', agentId: 'support-bot', appId: 'helpdesk' });

// Wait for completion
const result = await m.waitForJob(job_id);

// Semantic search
const results = await m.search('editor preferences', { userId: 'ali' });

// Agents, teams, webhooks — all supported
await m.runAgents({ agent: 'curator', autoFix: true });
const teams = await m.listTeams();

Full TypeScript types included (index.d.ts).

Memory Categories

Separate memory by user, agent, session, and application:

# User's personal memory
m.add(messages, user_id="ali")

# Specific agent's memory
m.add(messages, user_id="ali", agent_id="support-bot")

# Session-scoped memory
m.add(messages, user_id="ali", run_id="session-123")

# Application-scoped memory
m.add(messages, user_id="ali", app_id="helpdesk")

# Search filters by category too
m.search("preferences", user_id="ali", agent_id="support-bot")

Memory Agents

Three autonomous agents that analyze your memory:

🧹 Curator — Finds contradictions, stale facts, duplicates. Auto-cleans with auto_fix=True.

🔗 Connector — Discovers hidden connections, behavioral patterns, skill clusters.

📰 Digest — Weekly summary with headlines, trends, and recommendations.

Team Shared Memory

Share knowledge across your team:

POST /v1/teams {"name": "Backend Team"}
POST /v1/teams/join {"invite_code": "xK9m2Qw5ab"}
POST /v1/teams/3/share {"entity": "Redis"}

Search automatically includes shared team knowledge.

API Endpoints

Endpoint	Description
POST /v1/add	Add memories (async, returns job_id)
GET /v1/jobs/{id}	Check background job status
POST /v1/search	Semantic search
POST /v1/agents/run	Run memory agents
GET /v1/insights	AI-generated insights
POST /v1/teams	Create team
POST /v1/webhooks	Create webhook
GET /v1/keys	List API keys
POST /v1/keys	Create new API key
GET /v1/graph	Knowledge graph
GET /v1/timeline	Temporal search
GET /v1/stats	Usage statistics

Full docs: https://mengram.io/docs

Architecture

┌──────────────────────────────────────┐
│          Your AI Clients             │
│  Claude Desktop · Cursor · Windsurf  │
└──────────────┬───────────────────────┘
               │ MCP / REST API
┌──────────────▼───────────────────────┐
│        Mengram Cloud API             │
│  Extraction · Re-ranking · Search    │
├──────────────────────────────────────┤
│       Memory Agents Layer            │
│  🧹 Curator · 🔗 Connector · 📰 Digest│
├──────────────────────────────────────┤
│       Storage Layer                  │
│  PostgreSQL · pgvector · Teams       │
│  Webhooks · Reflections · Graph      │
└──────────────────────────────────────┘

License

MIT

---

Built by Ali Baizhanov